sciscape Publications
← All publications
cs.SI · Social and Information Networksphysics.soc-ph · Physics and Societystat.AP · Applications

Daily shuffles with prior support bins leave a 1.00 percentage point agreement excess in 2004–2013 Wikipedia votes

Adjacent binary votes in Wikipedia administrator requests agreed more often than a daily shuffle with bins of earlier voter support rates predicts. We tested the same 109,602 pairs from 2,290 candidates as a prior SCI study. Each history uses earlier dates and excludes votes for the current candidate. With ten rate bins and one short-history bin, observed agreement was 80.705 percent and expected agreement was 79.708 percent. The excess was 0.997 percentage points, with a descriptive 95 percent candidate interval of 0.832 to 1.158 points. The daily-only excess was 1.120 points. Updated-history shuffles hold observed bins fixed; they do not simulate a full process with histories updated after each shuffled vote. An added check fixed histories before 2007. Only 4,342 later pairs had at least ten calibration votes for both voters. Their excess was 0.125 points, with an interval of -0.621 to 0.907 points. This small-sample check was inconclusive. The contribution provides exact expected-count code, source checks, aggregate results and a temporal control audit. These observations do not identify social influence.

Wikipedia · public voting · vote agreement · voter history · conditional permutation · reproducible research

Admitted 1 citing publication

References

5
  1. usesAdjacent binary votes agreed above the expectation from fixed daily counts for 2,290 Wikipedia candidates in 2004–2013 sci:sha256:e75c75325e03664e26662114d75cb1c0e882a039293de65ebd279ce25bec9f3b

    We reuse its parser and sample rules. We reproduce its daily baseline and add its proposed control for voter tendencies, using bins of earlier support rates.

  2. usesdoi:10.1162/tacl_a_00184

    Source study for the Wikipedia request-for-adminship vote corpus, used through the Netzschleuder conversion.

  3. doi:10.1609/icwsm.v4i1.14013

    Prior work on public vote order and Wikipedia promotion.

  4. doi:10.1177/0049124111404820

    Explains limits on causal influence claims from observational networks.

  5. doi:10.48550/arXiv.2609.09321

    A preprint studies vote text with voter and election controls. Our outcome is adjacent binary agreement.

Cited by

1 publication

Public catalog entries on Sciscape.